SOMATIC CHROMOSOMES IN VICIA 3o5 



as Strasburger (o5), Dehorne (ii), and Mûller (12), believe they are 

 formed both by this method and through such adhésion as we hâve descri- 

 bed. It is quite probable that some of the anastomoses in Vicia may be ori- 

 ginale as - pseudopodia -, since they become so very numerous in later 

 stages. It seems clear from the figures, however, that the earlier ones are 

 formed for the most part by the adhérence ot two viscid bodies, as is 

 plainly the case with the similar structures occurring at anapliase, fig. 31. 



Reactions taking place between the chromosomes and the cytoplasm 

 and between the varions chromosomes now resuit in the production of the 

 nuclear cavity. The beginning of this process is shown in fig. 34, at the 

 left. The chromosome-cytoplasm reaction is hère revealed by the appea- 

 rance of small droplets of clear liquid between the two. As they increase 

 in size the chromosome becomes indented at thèse points, and the cyto- 

 plasm is pushed away into small concavities by the opposite side of the 

 droplets. The distinction between chromosome and cytoplasm is very clear. 

 The outer boundary of the droplets is not chromatic and remains colorless 

 even when the fine chromosomic anastomoses at somewhat later stages are 

 well stained. Thus the nuclear membrane, which is formed at the outer 

 side of the droplets, seems without doubt to be cytoplasmic, in contra- 

 distinction to the opinion of Vejdowsky (121, who holds that the small drop- 

 lets are really vacuoles within the chromosome, whose outer boundary is 

 consequently along their outer side. According to this view the nuclear 

 membrane represents the peripheral portion of certain chromosomes. In 

 this connection it is important to note that such droplets usuallyappearbefore 

 any vacuoles are visible in the interior; they hâve even been frequently 

 observed du ring anaphase along the margins of the chromosomes before 

 the latter hâve reached the pôles. The peculiar interprétation of Stomps 

 (10), namely that the nuclear vacuole originates by the association of small 

 preexisting vacuoles from the cytoplasm, has not found support in the pré- 

 sent study. 



The true telophasic vacuolization of the chromosomes begins at about 

 the time the latter begin to separate from one another, fig. 34-36. The 

 vacuoles first appear as obscure though rather sharply limited circular or 

 elongated spaces within the chromosome. They occur not only in the mé- 

 dian région but also near or against the periphery — in fact in almost every 

 conceivable position, as an inspection of the figures will show. This fact is 

 of the greatest importance, as it bears directly upon the question of the 



